similar to: VMI Interface Proposal Documentation for I386, Part 4

Fairly straightforward code motion in system.h into the sub-arch layer. Affected functionality include control register accessors, which are virtualizable but with great overhead due to the #GP cost; wbinvd, and most importantly, halt and interrupt control, which is non-virtualizable. Since read_cr4_safe can never fault on a VMI kernel (P5+ processor is required for VMI), we can omit the fault

Fairly straightforward code motion in system.h into the sub-arch layer. Affected functionality include control register accessors, which are virtualizable but with great overhead due to the #GP cost; wbinvd, and most importantly, halt and interrupt control, which is non-virtualizable. Since read_cr4_safe can never fault on a VMI kernel (P5+ processor is required for VMI), we can omit the fault

Illustration of how VMI inlines are used to greatly limit the impact of code change in low level assembler code. Spinlocks, system calls, and the fault handling paths are affected by adding some padding bytes to convert the native instructions into a hook point for the hypervisor to insert shim code. These changes are sufficient to glue the Linux low level entry points to hypervisor event

Illustration of how VMI inlines are used to greatly limit the impact of code change in low level assembler code. Spinlocks, system calls, and the fault handling paths are affected by adding some padding bytes to convert the native instructions into a hook point for the hypervisor to insert shim code. These changes are sufficient to glue the Linux low level entry points to hypervisor event

Index: linux-2.6.16-rc5/Documentation/vmi_spec.txt =================================================================== --- linux-2.6.16-rc5.orig/Documentation/vmi_spec.txt 2006-03-09 23:33:29.000000000 -0800 +++ linux-2.6.16-rc5/Documentation/vmi_spec.txt 2006-03-10 12:55:29.000000000 -0800 @@ -0,0 +1,2197 @@ + + Paravirtualization API Version 2.0 + + Zachary Amsden, Daniel Arai, Daniel Hecht,

Index: linux-2.6.16-rc5/Documentation/vmi_spec.txt =================================================================== --- linux-2.6.16-rc5.orig/Documentation/vmi_spec.txt 2006-03-09 23:33:29.000000000 -0800 +++ linux-2.6.16-rc5/Documentation/vmi_spec.txt 2006-03-10 12:55:29.000000000 -0800 @@ -0,0 +1,2197 @@ + + Paravirtualization API Version 2.0 + + Zachary Amsden, Daniel Arai, Daniel Hecht,

On 04/10/17 17:58, Josh Poimboeuf wrote: > Convert the hard-coded native patch assembly code strings to macros to > facilitate sharing common code between 32-bit and 64-bit. > > These macros will also be used by a future patch which requires the GCC > extended asm syntax of two '%' characters instead of one when specifying > a register name. > > Signed-off-by:

Convert the hard-coded native patch assembly code strings to macros to facilitate sharing common code between 32-bit and 64-bit. These macros will also be used by a future patch which requires the GCC extended asm syntax of two '%' characters instead of one when specifying a register name. Signed-off-by: Josh Poimboeuf <jpoimboe at redhat.com> ---

The new patching code means that we don't actually have to do as much work in lguest's patcher. We use paravirt_patch_default() for cases we don't want to patch inline, which automatically handles the "patch iret to use a direct jump" case. There's no measurable effect on lguest's virtbench scores tho. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> ---

There is now more than one place where we use the fact that bit 9 of eflags is the interrupt-enabled flag, so define EFLAGS_IF. We make it 512 so it can be used in asm, too. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> --- a/arch/i386/lguest/lguest.c +++ b/arch/i386/lguest/lguest.c @@ -107,9 +107,8 @@ static void fastcall irq_disable(void) static void fastcall irq_enable(void)

There is now more than one place where we use the fact that bit 9 of eflags is the interrupt-enabled flag, so define EFLAGS_IF. We make it 512 so it can be used in asm, too. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> --- a/arch/i386/lguest/lguest.c +++ b/arch/i386/lguest/lguest.c @@ -107,9 +107,8 @@ static void fastcall irq_disable(void) static void fastcall irq_enable(void)

The new patching code means that we don't actually have to do as much work in lguest's patcher. We use paravirt_patch_default() for cases we don't want to patch inline, which automatically handles the "patch iret to use a direct jump" case. There's no measurable effect on lguest's virtbench scores tho. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> ---

Fairly straight code motion. Split non-virtualizable pieces of processor.h into default and VMI subarches. CPUID is non-virtualizable, since it doesn't trap, and very often the hypervisor will want to hide specific feature bits from the kernel. To provide a replacement for call sites that use CPUID as a serializing instruction, the sync_core() macro is still available. Signed-off-by:

Fairly straight code motion. Split non-virtualizable pieces of processor.h into default and VMI subarches. CPUID is non-virtualizable, since it doesn't trap, and very often the hypervisor will want to hide specific feature bits from the kernel. To provide a replacement for call sites that use CPUID as a serializing instruction, the sync_core() macro is still available. Signed-off-by:

From: Boris Ostrovsky <boris.ostrovsky at oracle.com> commit 88c15ec90ff16880efab92b519436ee17b198477 upstream. As result of commit "x86/xen: Avoid fast syscall path for Xen PV guests", the irq_enable_sysexit pv op is not called by Xen PV guests anymore and since they were the only ones who used it we can safely remove it. Signed-off-by: Boris Ostrovsky <boris.ostrovsky at

Xen PV guests have been the only ones using it and now they don't. Signed-off-by: Boris Ostrovsky <boris.ostrovsky at oracle.com> --- arch/x86/entry/entry_32.S | 8 ++------ arch/x86/include/asm/paravirt.h | 7 ------- arch/x86/include/asm/paravirt_types.h | 9 --------- arch/x86/kernel/asm-offsets.c | 3 --- arch/x86/kernel/paravirt.c | 7

This is a note to let you know that I've just added the patch titled x86/paravirt: Remove the unused irq_enable_sysexit pv op to the 4.4-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary The filename of the patch is: x86-paravirt-remove-the-unused-irq_enable_sysexit-pv-op.patch and it can be found in the queue-4.4

(Andrew, please sit these in the -mm tree for cooking) Create a paravirt.h header for (almost) all the critical operations which need to be replaced with hypervisor calls. For the moment, this simply includes no_paravirt.h, where all the native implementations now live. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Zachary Amsden <zach@vmware.com>

(Andrew, please sit these in the -mm tree for cooking) Create a paravirt.h header for (almost) all the critical operations which need to be replaced with hypervisor calls. For the moment, this simply includes no_paravirt.h, where all the native implementations now live. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Zachary Amsden <zach@vmware.com>

plain text document attachment (xx-paravirt-irqs.patch) Interrupt updates for paravirt ops. Signed-off-by: Steven Rostedt srostedt@redhat.com Signed-off-by: Glauber de Oliveira Costa <gcosta@redhat.com> Index: clean-start/arch/x86_64/ia32/ia32entry.S =================================================================== --- clean-start.orig/arch/x86_64/ia32/ia32entry.S +++